An Architecture for Seamless Integration of UAS-based Wildfire - - PowerPoint PPT Presentation
An Architecture for Seamless Integration of UAS-based Wildfire Monitoring Missions UNIVERSITAT POLITCNICA DE CATALUNYA ICARUS Research Group Cristina.Barrado@upc.edu April'08 Outline Presentation of ICARUS Research Group
ICARUS Research Group Cristina.Barrado@upc.edu April'08
Remote Sensing, Salt Lake City, April 2008
Presentation of ICARUS Research Group Motivation System Technologies and Architecture Application Scenario: Wildfire Conclusions
Remote Sensing, Salt Lake City, April 2008
Technical University of Catalonia at
Barcelona
–
15 schools: EPSC
–
40 departments: DAC
–
30.000 students
–
2.500 PDI
Castelldefels School of Tech.
–
Electrical Engineering
–
Aeronautic Engineering
–
3.000 students
Computer Architecture Dep.
–
120 PDI
–
High Performance Computes (BSC)
–
Network Distributed Applications
Remote Sensing, Salt Lake City, April 2008
ICARUS – Intelligent Communications and Avionics for
E.Pastor (Ph.D.), C.Barrado (Ph.D.), M.A. Peña(Ph.D.), J.López, X.Prats, J.Ramírez, P.Royo E.Santamaria
Remote Sensing, Salt Lake City, April 2008
Computer Sciences – web services – embedded programming and compilers – GIS – formal methods and verification Electrical Engineering – WiFi, WiMax, RC, Satellite – Electronic board design Aeronautics Engineering – navigation – aeronavigation procedures – certification
Remote Sensing, Salt Lake City, April 2008
Remote Sensing, Salt Lake City, April 2008
Main resources
Remote Sensing, Salt Lake City, April 2008
Presentation of ICARUS Research Group Motivation System Architecture and technologies Application Scenario: Wildfire Conclusions
Remote Sensing, Salt Lake City, April 2008
State of the Art in application of UAS: – Firebird 2001: Fire Fighting Management
– ERAST / FiRE: NASA Project Design – WRAP: NASA / US Forest Service Project – Fire detection by Szendro Fire Department,
– NASA Dryden Flight Research Center
Remote Sensing, Salt Lake City, April 2008
MALAT Division of Israel Aircraft Industries – Demonstrated a system capable of fire
– Firebird: Payload 25 kg, endurance 5 h cruise 60
– Heron: Payload 250 kg, endurance 40 h cruise 80
Remote Sensing, Salt Lake City, April 2008
ERAST (Environmental Research Aircraft and
– Develop and flight-demonstrate UAVs for cost-
– ALTUS-II Payload 150 kg, endurance 12 h cruise 65
– ALTAIR scientific variant of the PREDATOR-B Payload 340 kg, endurance 32 h cruise 151
Remote Sensing, Salt Lake City, April 2008
WRAP (Wildfire Research and Applications
– Real fire monitoring missions over the USA
– Airborne InfraRed System (AIRDAS) – Thermal scan bands: 1 (0.61 - 0.68)
2 (1.57 -1.70) 3 (3.60 -5.50) 4 (5.50 -13.0)
– Calibration: IR +600 C. FOV: 108 degrees. Scan
Remote Sensing, Salt Lake City, April 2008
Small UAS used for early fire detection: – Low cost, simple approach – Fire department integrated UAV
Remote Sensing, Salt Lake City, April 2008
Earth Observation and the Role of UAVs, a
Remote Sensing, Salt Lake City, April 2008
Remote Sensing, Salt Lake City, April 2008
Presentation of ICARUS Research Group Motivation System Technologies and Architecture Application Scenario: Wildfire Conclusions
Remote Sensing, Salt Lake City, April 2008
Firemen requirements: GRAF System Architecture Communication Gateway Mission: HMI and End User Procedures
Remote Sensing, Salt Lake City, April 2008
decentralized by regions.
cooperation available if necessary. Area: 31 932 km2 Population: 6.704.146 Fires during 2006: 629 Burnt area: 3.404 ha Worst year (1994): 76.125 ha
Remote Sensing, Salt Lake City, April 2008
Remote Sensing, Salt Lake City, April 2008
Surveillance and attack
In case of detection first
Rest of available units are used
No flying during night time.
Remote Sensing, Salt Lake City, April 2008
Geographical application area:
– Relatively small area; operations under responsibility of local
government and therefore with limited budget.
– Externalized aerial resources except C&C helicopters. – UAS to be operated by external providers.
Integration with fire fighters own systems:
– Aerial operators see opportunities but do not want to see a
UAS mixed in their airspace!!
– Ground firefighters are eager to receive any available
technology innovation.
– Even though existing legal limitations and pilots opposition,
ground firefighters suggest several application scenarios with strict manned/unmanned separation.
Remote Sensing, Salt Lake City, April 2008
Identify effective application scenarios
– Contacts with many fire fighter organizations – Application scenarios change depending on user
– Human-Machine Interface critical for non-IT users
Identify operational and information flow and
– Highly dependent on the selected autopilot – Information flow – management – exploitation: key
Remote Sensing, Salt Lake City, April 2008
Oriented to mission management and information
Real Time Data Acquired and Distribution System divided into four components:
– UAS: designed for data acquisition and autonomous
– Mobile Control Station: responsible for UAS tactical control
(flight operations), data gathering and processing.
– Squad Information Terminal: provides information to the
ground crew.
– Data Processing Center: strategic control of multiple
level coordination and decision center.
Remote Sensing, Salt Lake City, April 2008
UAS components: – Airframe – Flight Control System – Payload – Payload/Mission
– Communication System
Remote Sensing, Salt Lake City, April 2008
Network Centric with data Publish/Subscribe Services may be producers and/or consumers
Remote Sensing, Salt Lake City, April 2008
Remote Sensing, Salt Lake City, April 2008
Remote Sensing, Salt Lake City, April 2008
Remote Sensing, Salt Lake City, April 2008
makes service location is irrelevant monitors links to provide cost effective QoS
Remote Sensing, Salt Lake City, April 2008
Mission is formally specified through visual tools:
– Relations between services are specified by flow diagrams – Dynamic activities through event-based systems.
Remote Sensing, Salt Lake City, April 2008
Dynamic service discover
– Services can be identified when the system goes online or
later during operation. Remote execution
– Consumer simply sends a service request and its parameters.
Later on it will get results. Self-description
– Each module provides a description of the services that it can
Multiple equivalent services may be available adding a level
Data streaming
– Semantic publish/subscription mechanisms for high change
rate data
Remote Sensing, Salt Lake City, April 2008
The mission is a set of services that orchestrate
Link the flight plan that the UAS follows and the
Mission may dynamically change as fire evolves,
Given that operational requirements change from
Remote Sensing, Salt Lake City, April 2008
Previous to flight: mission definition During flight (in parallel): – Exploration (flight area redefinition) – Data processing – Data Presentation – Data Storage After flight: Data post-processing
Remote Sensing, Salt Lake City, April 2008
Previous to flight: mission definition During flight (in parallel): – Exploration (flight area redefinition) – Data processing – Data Presentation – Data Storage After flight: Data post-processing
Remote Sensing, Salt Lake City, April 2008
Responsible: Decision maker – Data Presentation Mission Operator – Definition – Exploration Others – Pilot-in-command – Maintenance
Remote Sensing, Salt Lake City, April 2008
Continuous scan is necessary to follow fire perimeter.
Remote Sensing, Salt Lake City, April 2008
Mission control allows to design a service that identifies fire perimeter Exploration area is dynamically changed by updating a few flight plan parameters.
Remote Sensing, Salt Lake City, April 2008
Remote Sensing, Salt Lake City, April 2008
Dynamic adaptation
Remote Sensing, Salt Lake City, April 2008
Dynamic adaptation
Remote Sensing, Salt Lake City, April 2008
Dynamic adaptation
Remote Sensing, Salt Lake City, April 2008
Remote Sensing, Salt Lake City, April 2008
Presentation of ICARUS Research Group Motivation System Technologies and Architecture Application Scenario Conclusions
Remote Sensing, Salt Lake City, April 2008
GRAF identified three viable fire missions: – Monitoring of prescribed burnings for security
– Final fire mop-up with hot-spots detection – Fire monitoring during night Based on the following characteristics: – Detection is not a goal in populated areas – No interference with standard air-planes – High cost of alternate solutions – Progressive complexity
Remote Sensing, Salt Lake City, April 2008
Mean of burned surface per wildfire is 6Ha FLIR A320 camera characteristics: – 320 x 200 pixels – At 100 AGL: image is 40 x 30 m
– Frequency rate up to 9Hz UAV flight characteristics – 100 meters AGL – speed 60Km/h
Remote Sensing, Salt Lake City, April 2008
ICARUS group research topics and proposals for
– Hardware/Software embedded architecture – Human-Machine Interfaces – End Users Requirements and Procedures – Integration of UAV in air space (future) Forest fire is an target civil application for: – Segregated airspace – UAS operation is cost effective – Commercially viable
Remote Sensing, Salt Lake City, April 2008
Not all geographical scenarios are equivalent;
Other applications (fire perimeter detection, crop
Remote Sensing, Salt Lake City, April 2008